Can courses be autotelic?

This is a bit of follow up to my previous post.  I mentioned having a sense of deja vu in reading psychology books, which is true conceptually but doesn't hold for the terminology.  Autotelic, for example, is a new word for me.  The New Oxford American Dictionary built into the Kindle software says it means having a purpose in and of itself.  If a course were autotelic, what would it look like?

When I finished my undergrad studies, I hung around Ithaca for another semester and did a bunch of different small things that together occupied me.  One of those things was to go to a Political Science lecture.  I had taken a fair number of upper level courses, in Poli Sci, but I missed the principles classes, so this was going back to fundamentals.  I had heard good things about the instructor, Isaac Kramnick, and a girl I was kind of crazy about at the time was also interested.  We attended together.  We didn't do the readings or the assignments, but if memory serves we also didn't miss a class.  This was an attraction not unlike good foreign films, which were compelling.  Perhaps an autotelic course has other aspects, but shouldn't attractiveness of this sort be at least a part?

This morning I had a coffee with Jim, which we do once in a while, and we got to talk a little about Econ courses online and whether having a famous economist as the "teacher" in the course mattered or if that were trumped by the course materials being well designed.  Put a different way, among the leading principles texts on the market, one is authored by Paul Krugman and his wife Robin Wells.  Krugman has his column in the NY Times every Monday and Friday, and he seems to be on TV a lot, on Charlie Rose, The Daily Show, other stuff as well.  It is not hard to imagine that most undergraduates know of Krugman, whether they've taken an economics course or not.  Another leading textbook is by Greg Mankiw.  Mankiw is pretty well known too.  He was Chairman of the Council of Economics Advisers during the term of Bush II. 

Now ask the following simple question (though the answer is not transparent at all).  Suppose I teach a  principles course using either Krugman's or Mankiw's text.  Among other things, this means I write the exams, so the tests are original for that particular offering of the class.  Now suppose there is a lot of online video content.  Should that video content feature one of these superstars?  Or should it feature me?  In other words, does star power matter more or does it matter more that the author of the exams be the presenter?  Then ask whether either way, is the course autotelic?  Or do autotelic courses not have exams?

My university has a requirement that every course must have a final exam.  I suspect that many other places have similar requirements.   I violated that requirement in a course I taught a year ago.  I wasn't trying to shirk at all, but there were so few students I taught the class as a seminar and the students did enough other activities to be assess that I thought an exam was redundant and perhaps pernicious.  There is much teaching to the test especially in the high enrollment classes.  That seems the opposite of a what an autotelic class would be like. 

George Kuh, pretty clever guy in my view, use the term "engagement" in talking about well functioning college instruction and NSSE is an instrument aimed at measuring that across the entire college experience.  I haven't tracked it recently but in the early years of that survey the findings were disarming - students weren't engaged much at all.  If memory serves, some of NSSE concerns an undergraduate research experience, which likely would be outside a course setting.  If a college student works in a professor's lab, that experience might very well be autotelic.  And that's great, but it still leaves us with the question, what about courses?

Suppose we concluded that courses can be autotelic and that star power isn't the key ingredient.  Rather it is making the course challenging to the student but with the right sort of feedback where the student sees that by intensive effort the student can do well and sees how what is learned can be well applied outside the course setting.  (I'm pretty much lifting this directly from Flow, but then translating it into an instructional setting.)    The course design is not a guarantee that students will experience flow in taking the course.  If they are hell bent on being disengaged, that will trump the course design.  But the design should be reasonably effective on students who try it earnestly.  Again, this is a supposition so I can pose the next question.  If it's possible, then would it become a mandate to produce autotelic courses? 

It certainly doesn't seem there is such a mandate now.  Should we infer from that that people think it impossible to make an autotelic class?  Or have folks simply not taken up the question? If not, why not?